HSE Health & Safety Executive Seismic hazard: UK continental shelf Prepared by EQE International Ltd for the Health and Safety Executive OFFSHORE TECHNOLOGY REPORT 2002/005 HSE Health & Safety Executive Seismic hazard: UK continental shelf EQE International Ltd The Beacons Warrington Road Birchwood Cheshire WA3 6WJ United Kingdom HSE BOOKS © Crown copyright 2002 Applications for reproduction should be made in writing to: Copyright Unit, Her Majesty’s Stationery Office, St Clements House, 2-16 Colegate, Norwich NR3 1BQ First published 2002 ISBN 0 7176 2335 1 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means (electronic, mechanical, photocopying, recording or otherwise) without the prior written permission of the copyright owner. This report is made available by the Health and Safety Executive as part of a series of reports of work which has been supported by funds provided by the Executive. Neither the Executive, nor the contractors concerned assume any liability for the reports nor do they necessarily reflect the views or policy of the Executive. ii FOREWORD This report on seismic hazard mapping of offshore Britain is a summary of work carried out by EQE International Limited for the Health and Safety Executive (HSE). This work was undertaken in conjunction with NORSAR, Oslo, who have at the same time been working on the seismic zonation for Norway, on behalf of the Norwegian Council for Building Standardization (NBR). Through the coordination and synchronization of the British and Norwegian seismic hazard mapping projects:- a degree of harmonization which has not previously been achieved, the results of these studies provide an internationally consistent basis for offshore seismic loading in the most seismically active area offshore Britain. The contents of the report are the work of EQE International Limited, and do not necessarily reflect the policies of the HSE. iii iv EXECUTIVE SUMMARY This report documents an investigation of seismic hazard in UK offshore waters, which has been conducted by EQE with the scientific collaboration of the Norwegian Seismic Array (NORSAR). The results are expressed probabilistically, and displayed graphically in a series of contour maps of peak ground acceleration for return periods of 100 years, 200 years, 495 years, 1000 years, and 10,000 years. The northern North Sea region has the highest level of seismic hazard in UK offshore waters. Whereas previous seismic hazard maps for the North Sea have shown discontinuities in hazard levels across the boundary separating the British and Norwegian sectors, the hazard maps produced in this joint Anglo-Norwegian study satisfy the condition of continuity across the sector boundary. This consistency is achieved through agreement on unified seismic source and ground motion attenuation models in the northern North Sea, and represents a major advance in seismic hazard assessment across international frontiers. The harmonized Anglo-Norwegian seismic hazard maps show that the highest peak ground acceleration hazard in UK offshore waters is attained in the northern North Sea. Close to the sector boundary, the 10-4/yr exceedance peak ground acceleration can reach values of 30%g. The seismic hazard is somewhat less in the southern North Sea, where the 10-4/yr exceedance peak ground acceleration can reach values of almost 25%g. Outside these two specific regions, the hazard is lower near the Western UK coast, typically about 20%g offshore Wales and Northwest England; and the hazard is smaller elsewhere. For the specification of bedrock earthquake loading at any offshore site, a common EQE- NORSAR approach has been formulated. This involves the specification of a single seismic response spectral shape, which is anchored at 40Hz to the relevant site-specific peak ground acceleration for the requisite return period. v vi CONTENTS EXECUTIVE SUMMARY iii 1. INTRODUCTION 1 2. THE SPATIAL PATTERN OF SEISMICITY 5 2.1 Historical Earthquakes 5 2.2 Instrumental Seismological Data 6 2.3 Seismotectonics 7 2.4 Neotectonics 10 3. SEISMIC HAZARD SOURCE MODEL 11 3.1 Seismic Hazard Computation 11 3.2 Seismic Area Zonation 13 3.3 Seismic Activity Rate Distribution 15 3.4 Maximum Magnitude Distribution 16 3.5 b-Value Distribution 18 4. SEISMIC GROUND MOTION 19 4.1 Choice of Attenuation Relations 19 4.2 The Attenuation Relations of Ambraseys, Simpson and Bommer 20 4.3 The Attenuation Relations of Toro, Abrahamson and Schneider 21 5. SEISMIC HAZARD COMPUTATION 23 5.1 Regional Acceleration Hazard 23 5.2 Seismic Response Spectra 24 5.3 Earthquake Time Histories 25 5.4 Soil Response Effects 26 5.5 Conclusions 27 ACKNOWLEDGEMENTS 27 REFERENCES 29 TECHNICAL GLOSSARY 33 vii FIGURES Figure 1.1: Map of the structural framework of the UK and the Norwegian North Sea Figure 2.1: Known N.W. European epicentres: Magnitudes m 4 MS or equivalent Figure 2.2: N.W. European epicentres reported by ISC, 1904 - 1990, Magnitudes m 3 Figure 2.3: BGS offshore epicentres: 1990 - 1990 inclusive Magnitudes m 2 ML Figure 3.1: Logic-tree branches for representing modelling parametric uncertainty Figure 3.2: Zonation model “A”, interfaced to NORSAR model 1 Figure 3.3: Zonation model “B”, interfaced to NORSAR model 2 Figure 5.1: Peak ground acceleration contours (m/s2) for an annual exceedance probability of 1E-2 Figure 5.2: Peak ground acceleration contours (m/s2) for an annual exceedance probability of 5E-2 Figure 5.3: Peak ground acceleration contours (m/s2) for an annual exceedance probability of 2.1E-2 Figure 5.4: Peak ground acceleration contours (m/s2) for an annual exceedance probability of 1E-3 Figure 5.5: Peak ground acceleration contours (m/s2) for an annual exceedance probability of 1E-4 Figure 5.6: Comparison of spectral shapes at location [1.5E, 58 N] for annual exceedance probabilities of 1E-2, 1E-3 and 1E-4 Figure 5.7: Comparison of generic offshore spectral shape with 1981 PML hard ground spectra normalized to 1g Figure 5.8: Comparison of generic offshore spectral shape with 1988 PML hard ground uniform risk spectral shape Figure 5.9: Ratio between the shaking in bedrock and on top of soil deposits for probabilities of exceedance of 1E-4 [a] and 1E-2 [b]. viii APPENDIX 1: 37 First Seismic Zonation APPENDIX 2: 47 Second Seismic Zonation ix x 1. INTRODUCTION 1.1 CROSS-FRONTIER COLLABORATION The methodologies adopted to evaluate seismic hazard may vary significantly from country to country, even where the countries share a common border and the regional seismic zones overlap. Apart from methodological differences, the underlying seismological and geological databases maintained in neighbouring countries may also differ substantially in scope, reliability and interpretation, and thus cause further international disparities in the conduct and output of seismic hazard assessment. For reasons both technical and administrative, it is rare for a collaborative industrial venture to be established, involving independent organizations from several countries, which is aimed at joint probabilistic seismic hazard mapping for a region spanning a common international border. All too often, hazard contour maps display a discontinuity across frontiers, which is explained by an international breakdown in human communication rather than being attributable to any natural geological fracture. The comparative study reported herein was a collaborative and cooperative endeavour combining the seismic hazard expertise available to EQE and NORSAR, and involved a transparent exchange programme of knowledge, information and expert judgement. This has been made contractually feasible through the synchronization of the work done for offshore Britain, under contract to the Health and Safety Executive (HSE), with that carried out by NORSAR for offshore Norway, under a contract to the Norwegian Council for Building Standardization (NBR). This latter project (NORSAR, 1998), has been sponsored by various Norwegian government and petroleum organizations: The National Fund for Natural Disaster Assistance; The National Office of Building Technology and Administration; Norwegian Water Resources and Energy Administration; A/S Norske Shell; Norsk Hydro; Saga Petroleum; and Statoil A/S. This Norwegian project has also involved technical cooperation with NGI, who collaborated with EQE and NORSAR on the joint study of Earthquake Loading on the Norwegian Continental Shelf (Bungum and Selnes, 1988). This study was fully documented in fifteen specialized technical reports. Broader in geographical perspective than ELOCS, the prime goal of the current Norwegian project has been the development of a seismic zonation for Norway, resulting in the production of seismic hazard maps for Norwegian onland and offshore areas that can form part of the Norwegian National Application Document under Eurocode 8. The benefits of a cross-frontier collaboration for seismic hazard assessment are manifold: rationalisation of scientific databases; reconciliation of hazard evaluation procedures; pooling of expert judgement; sharing of internal documentation; independent international validation of hazard computation etc.. Although this is more than a decade away from being the first UK study of North Sea seismic hazard, (see e.g. Woo and Muir Wood, 1986), this is the first study to which these important benefits accrue. Given the substantial amount of common work shared between EQE and NORSAR, and the open publication of the extensively detailed NORSAR (1998) report, some of the common documentation is not duplicated here. Instead, copious reference is made to the NORSAR report, which is publically available, and which is recommended to readers wishing to gain a more complete international picture of technical aspects of this Anglo-Norwegian collaboration. 1 Apart from transcending international frontiers, another original facet of this study, which sets it apart from predecessors, is that this is the first study of UK offshore seismic hazard which incorporates the logic-tree methodology (Kulkarni et al., 1984) for representing parametric uncertainty. Traditionally, single best-estimate values are assigned to input model parameters.